CODEWITHSUNDEEP
ssisetllookupstar-schemafact-table
John_Rodgers
John_Rodgers

Reputation: 181

SSIS - Incremental Load - Lookup Object

I am building a SSIS project in order to create an ETL to load a Data Warehouse that uses a star schema model.

I've 3 dimensions:

 - Customer (Customer_ID, Customer_Name, Email)
 - Employees (Employee_ID, Employee_Name, Salary)
 - Product (Product_ID, Product_Category)

And my Fact Tables is very simples, just:

- Customer_ID
- Employee_ID
- Product_ID
- Sales_Amount

I'm trying to create a package using SSIS. I see a lot of webinars in order to learn how to make incremental load using SSIS and many of them suggest me that use the following objects structure: enter image description here

In OLE DB Source I've the following query:

SELECT a.Customer_ID,
       b.Employee_ID,
       c.Product_ID,
       sa.Sale_Amount
FROM Staging_Area_table sa
LEFT JOIN Customer a ON
sa.Customer_Name = a.Customer_Name
LEFT JOIN Employee b ON
sa.Employee_Name = b.Employee_Name
LEFT JOIN Product c ON
sa.Product_Category = c.Product_Category

My questions is: 1) It gives me an error :) That one:

[Lookup [61]] Error: Row yielded no match during lookup.

2) If I want the rows that doens't match in Fact Table why I'm putting "Lookup Match Output" during the lookup objects

3) This is a good approach to load my fact table?

Thanks!!!!!

Upvotes: 1

Views: 628

Answers (1)

sam
sam

Reputation: 1294

Answer to your first question -

You are getting the error because fail component is selected in lookup transaformation in the general tab. So you are getting error because there are no matched rows. To ignore the error select "Redirect rows to nomatch output in the general tab of lookup.

Second question - Many lookups are been needed because we have to make sure that all the dimentions are present before loading in fact tables.

Third Question - Yes this is the approach that many people follow.

Upvotes: 2

Related Questions